Skip to main content

Table 3 Models’ performance based on different datasets

From: Can a deep learning model based on intraoperative time-series monitoring data predict post-hysterectomy quality of recovery?

 

Accuracya

Sensitivitya

Specificitya

F1 scorea

AUROC

Preoperative data

 Deep learning

0.61 (0.57–0.65)

0.61 (0.54–0.68)

0.61 (0.50–0.71)

0.60 (0.56–0.64)

0.65 (0.62–0.67)

 Logistic regression

0.63 (0.59–0.66)

0.62 (0.59–0.65)

0.63 (0.56–0.71)

0.62 (0.60–0.65)

0.68 (0.66–0.70)

 Support vector machine

0.61 (0.56–0.66)

0.51 (0.40–0.62)

0.70 (0.59–0.81)

0.56 (0.49–0.63)

0.65 (0.60–0.70)

 Random forest

0.62 (0.60–0.65)

0.59 (0.49–0.70)

0.66 (0.59–0.72)

0.60 (0.55–0.66)

0.68 (0.65–0.70)

Intraoperative intervention data

 Deep learning

0.74 (0.70–0.79)

0.73 (0.66–0.80)

0.74 (0.61–0.87)

0.74 (0.71–0.77)

0.79 (0.75–0.82)

 Logistic regression

0.76 (0.71–0.81)

0.77 (0.73–0.80)

0.76 (0.64–0.88)

0.76 (0.73–0.79)

0.78 (0.74–0.82)

 Support vector machine

0.59 (0.54–0.64)

0.50 (0.41–0.59)

0.67 (0.55–0.80)

0.54 (0.48–0.60)

0.65 (0.61–0.68)

 Random forest

0.73 (0.67–0.79)

0.75 (0.71–0.78)

0.72 (0.58–0.86)

0.74 (0.69–0.78)

0.81 (0.76–0.85)

Intraoperative monitoring data

 Deep learningb

0.70 (0.69–0.72)

0.64 (0.58–0.69)

0.77 (0.70–0.84)

0.68 (0.66–0.71)

0.77 (0.72–0.81)

 Logistic regressionc

0.69 (0.63–0.75)

0.68 (0.64–0.72)

0.69 (0.56–0.83)

0.68 (0.64–0.73)

0.72 (0.68–0.77)

 Support vector machinec

0.62 (0.58–0.66)

0.61 (0.56–0.66)

0.63 (0.52–0.75)

0.62 (0.59–0.64)

0.68 (0.65–0.71)

 Random forestc

0.61 (0.57–0.65)

0.83 (0.72–0.94)

0.40 (0.21–0.58)

0.68 (0.66–0.69)

0.74 (0.73–0.76)

Intraoperative monitoring data + SmtO2

 Deep learningb

0.71 (0.69–0.73)

0.64 (0.57–0.72)

0.77 (0.68–0.87)

0.69 (0.67–0.70)

0.77 (0.74–0.79)

 Logistic regressionc

0.69 (0.63–0.75)

0.68 (0.64–0.72)

0.69 (0.54–0.85)

0.69 (0.65–0.72)

0.73 (0.68–0.78)

 Support vector machinec

0.67 (0.64–0.70)

0.63 (0.57–0.69)

0.70 (0.61–0.79)

0.65 (0.63–0.68)

0.71 (0.67–0.76)

 Random forestc

0.65 (0.60–0.70)

0.87 (0.79–0.95)

0.44 (0.27–0.61)

0.71 (0.70–0.73)

0.78 (0.73–0.82)

Preoperative data + intraoperative monitoring data + intraoperative intervention data

 Deep learning

0.73 (0.70–0.76)

0.74 (0.69–0.80)

0.71 (0.62–0.80)

0.73 (0.71–0.75)

0.81 (0.78–0.83)

 Logistic regression

0.73 (0.66–0.80)

0.75 (0.70–0.80)

0.72 (0.58–0.85)

0.74 (0.68–0.79)

0.77 (0.70–0.85)

 Support vector machine

0.59 (0.56–0.61)

0.50 (0.40–0.60)

0.67 (0.56–0.78)

0.54 (0.49–0.59)

0.65 (0.61–0.69)

 Random forest

0.76 (0.72–0.80)

0.82 (0.75-0.88)

0.70 (0.57–0.83)

0.77 (0.74–0.80)

0.82 (0.78–0.87)

  1. Data are presented as mean (95% confidence interval)
  2. AUROC area under the receiver operating characteristic curve, SmtO2 muscular tissue oxygen saturation
  3. aCalculated based on the decision threshold of 0.5
  4. bBased on time-series data
  5. cBased on the maximum, minimum, mean, and standard deviation values of time-series data
Back to article page

Perioperative Medicine

ISSN: 2047-0525